machine learning

Open Climate Fix says its Quartz Solar tool saves Great Britain’s grid operator GBP 30 million ($39 million) per year through more accurate forecasting, which reduces the reserve capacity needed for balancing the electricity system. The non-profit company has used machine learning techniques with satellite, weather and historical generation data to reduce forecasting errors by half.

The Dutch company says its short-term trading solutions for solar and other renewable energy technologies, supports grid balancing, reduces the costs of imbalance, and optimizes energy flows in an “increasingly volatile” energy market. It is growing internationally, and expanding its support of battery-related trading.

Scientists have developed a new model for heat exchangers of heat pumps, combining strengths of numerical modeling and machine learning.

Using Google Earth imagery and 2019-2022 Sentinel-2 datasets, Chinese scientists have developed a two-stage classification framework to obtain the annual global dataset of solar photovoltaic panels at 20-meter resolution from 2019 to 2022.

Researchers in Saudi Arabia have compared the performance of ground-mounted PV plants with that of off-shore solar facilities and have found that floating installations benefit from the cooling effect of the seawater.

Scientists have created a novel probabilistic model for 5-minutes ahead PV power forecasting. The method combines a convolutional neural network with bidirectional long short-term memory, attention mechanism, and natural gradient boosting.

An international research team has developed a novel approach for predicting inverter temperature through symbolic regression based on particle swarm optimization.

Scientists in Spain have used genetic algorithms to optimize a feedforward artificial neural network for the prediction of energy generation of PV systems. Genetic algorithms use “parents” and “offspring” solutions to achieve better results in subsequent generations.”

An international team has combined organic synthesis with predictive models to discover new functional materials that enhance performance of hole transport layers used in perovskite solar cells. The team asserts that optimizing for other solar cell properties is possible with the platform, as well as using it for development of materials for other kinds of devices.

The PV industry is embracing artificial intelligence and machine learning (ML) techniques to automate operations and maintenance (O&M) diagnostics and predictive analytics in PV systems. More transparency and standard definitions are needed, however, as US-based Sandia Labs scientists Joshua Stein and Marios Theristis explain.